Strain surveillance during chemotherapy to improve cardiovascular outcomes: the SUCCOUR-MRI trial

Abstract

Background and aims: The detection of cancer therapy-related cardiac dysfunction (CTRCD) by reduction of left ventricular ejection fraction (LVEF) during chemotherapy usually triggers the initiation of cardioprotective therapy. This study addressed whether the same approach should be applied to patients with worsening of global longitudinal strain (GLS) without attaining thresholds of LVEF.

Methods: Strain surveillance during chemotherapy for improving cardiovascular outcomes (SUCCOUR-MRI) was a prospective multicentre randomized controlled trial involving 14 sites. Of 355 patients receiving anthracyclines with normal baseline LVEF, 333 patients (age 59 ± 13 years, 79% women) with at least one other CTRCD risk factor, able to undergo magnetic resonance imaging (MRI), GLS, and three-dimensional echocardiography were tracked over 12 months. A total of 105 patients (age 59 ± 13 years, 75% women, 69% breast cancer) developing GLS-CTRCD (>12% relative reduction of GLS without a change in LVEF) were randomized to cardioprotection with neurohormonal antagonists vs. usual care. The primary endpoint was 12-month change in MRI-LVEF; the secondary endpoint was MRI-LVEF-defined CTRCD.

Results: During follow-up, two patients died, and two developed heart failure. Most patients were randomized at 3 months (62%). Median doses of angiotensin inhibition/blockade and beta-blockade were 75% and 50% of respective targets; 21 (43%) had side-effects attributed to cardioprotection. Due to a smaller LVEF change from baseline with cardioprotection than usual care (-2.5 ± 5.4% vs. -5.6 ± 5.9%, P = .009), follow-up LVEF was higher after cardioprotection (59 ± 5% vs. 55 ± 6%, P < .0001). After adjustment for baseline LVEF, the mean (95% confidence interval) difference in the change in LVEF between the two groups was -3.6% (-1.8% to -5.5%, P < .001). After cardioprotection, 1/49 patients developed 12-month LVEF-CTRCD, compared to 6/56 in usual care (P = .075). Global longitudinal strain improved at 3 months post-randomization in the cardioprotection group, with little change with usual care.

Conclusions: In patients with isolated GLS reduction after anthracyclines, cardioprotection is associated with better preservation of 12-month MRI-LVEF compared with usual care.

Keywords: Cancer therapy-related cardiac dysfunction; Global longitudinal strain; Left ventricular ejection fraction.

Structured Graphical Abstract

From 355 patients with normal left ventricular function undergoing anthracycline therapy, 333 with additional risk factors for cancer therapy-related cardiac dysfunction (CTRCD) were able to undergo baseline and magnetic resonance imaging and echocardiography. Of these, 105 developed isolated impairment of global longitudinal strain (GLS), and were randomized to cardioprotection with neurohormonal inhibition or usual care. Patients taking cardioprotection experienced a smaller change in left ventricular ejection fraction (LVEF) over 12 months. The baseline-corrected mean difference in LVEF was −3.6%. Normal LVEF and GLS = not meeting an LVEF-CTRCD criteria or a GLS change >12%; GLS only = an isolated >12% relative reduction in GLS; Lost = patients lost to follow-up. EF = patients who met the LVEF-CTRCD criteria.

Figure 1

Left ventricular responses to potentially cardiotoxic chemotherapy. CTRCD, cancer-treatment-related cardiac dysfunction; CTRCD, cancer-treatment-related cardiac dysfunction; EF+, attaining threshold for CTRCD on ejection fraction criteria; EF−, not attaining threshold for CTRCD on ejection fraction criteria; GLS+, attaining threshold for CTRCD on global longitudinal strain criteria; GLS−, not attaining threshold for CTRCD on global longitudinal strain criteria

Figure 2

Progression of subclinical left ventricular dysfunction (impaired global longitudinal strain with preserved ejection fraction) with and without cardioprotection. The patient on cardioprotection has preserved ejection fraction at the end of follow-up, with a 3.6% drop. The patient without cardioprotection has a 6.3% deterioration of ejection fraction, almost to the abnormal range

Figure 3

Baseline and 12-month follow-up left ventricular ejection fraction in cardioprotection and usual care groups. Baseline MRI-LVEF was similar in both groups (61 ± 5% vs. 60 ± 6%, P = .30), but cardioprotection was associated with better preservation of follow-up left ventricular ejection fraction than usual care. Note that study entry was based upon three-dimensional echocardiographic ejection fraction >50%, but baseline magnetic resonance imaging was ≤50% in 1 patient

Figure 4

Evolution of global longitudinal strain in patients randomized to cardioprotection and usual care. Individual data-point, mean and standard deviation show that the reduction in global longitudinal strain between the start of echo surveillance (baseline) and the time of randomization was similar in both groups. However, the trajectory after 3, 6, and 9 months post-randomization (R + 3, R + 6, R + 9) was different, with recovery of global longitudinal strain in the cardioprotection group, and minimal change in the usual care group. The accompanying table shows within and between-group differences, as well as numbers of patients. Data are sparse at 9 months post-randomization (22 in cardioprotection and 14 in usual care). The P values compare the current visit vs. the prior visit